Progressive Scottish Opinion have produced a second poll for the Glasgow East by-election. As with ICM, it shows a substantial Labour lead with voting intentions of LAB 52%, SNP 35%, CON 7%, LDEM 3%. The poll was conducted between the 14th and 17th July, and interviewed 509 people.

Progressive Scottish Opinion are not registered with the BPC, so they are not obliged to publish results. Some people get very sniffy about pollsters who aren’t registered with the BPC. It should be pointed out that membership of the BPC isn’t compulsory – it’s a voluntary agreement between the pollsters who produce high profiles polls for the media agreeing to publish their tables and answer questions. Companies outside it are often perfectly reputatable companies, often highly experienced, working inside MRS rules and codes of conduct. There’s nothing stopping them releasing their tables either if they wish to (hint!)

The downside is we don’t know things such as how the poll dealt with turnout, or don’t knows, what weighting they used and so on. We don’t though if they faced the same problems ICM did getting a representative sample. Thus while it seems likely that this poll is backing up ICM’s findings and showing a substantial lead, we can’t be certain that it isn’t just using some form of methodology that is more generous to Labour.

6 Responses to “Another poll shows Labour ahead in Glasgow East”

  1. Media reports have the undecided in this poll at 29%, a very large amount this close to the vote. Not a good sign for Labour in one of their strongholds. I would assume most of that vote is former Labour voters trying to decide whether to hold their nose and vote Labour again or make the leap to another party. Also it does not bode well for a large turnout as many may decide to stay home. In a riding with historically low turnouts to start with, this is another problem for an incumbent party.

    By extrapolition, the percentages are taken from 361 decided responders (at most, with no mention of those refusing to state) thus increasing the margin of error.

    Without the raw data or the methodology, not much can be taken from this poll except for the 29% undecided figure.

  2. Brian, we can’t even be that sure about the undecideds. The reports just say that 29% haven’t made their minds up yet – that could be 29% don’t knows, which would seem very high – or it could be a separate question asking if people would definitely vote for the party they said or whether they might change their mind, in which case it wouldn’t be particularly high at all!

  3. If the 29% figure includes those who may change their mind, then the original percentages attributed to the party support would seem even more suspect.

  4. For all the reservations about this poll, 17% is a big lead to explain away.

  5. A report that both SNP and Labour private polls suggest the contest is very close:
    I’m always very wary and chary about ‘leaks’ from parties’ ‘internal polling’.
    For a start, I doubt whether this means commissioned and properly carried out professional polls. At Harris and Wirthlin Europe) we did ‘private polling’ for the Conservatives for around ten year (in my time) and we never did poll in byelections. Any figure are likely to be based on the notoriously unreliable canvass returns.
    Second, parties are in the business of trying to win elections – fair enough – and so selective releases of information will be designed to that end. The SNP want themselves to be portrayed a having a better chance than the two published polls. Labour is relying on a high turnout from their traditionl support base in Glasgow East. It is in the best interests of both parties to encourage the idea that the contest is very close.
    However, all the above does not detract from the strong possibility that it really will be much closer on Thursday than ‘teen per cent’ suggested by two polls facing all kinds of difficulties in their design and execution.